Weak Interactions in Protein Folding: Hydrophobic Free Energy, van der Waals Interactions, Peptide Hydrogen Bonds, and Peptide Solvation

Hydrophobic free energy has been widely accepted as a major force driving protein folding [1, 2], although a dispute over its proper definition earlier made this issue controversial. When a hydrocarbon solute is transferred from water to a nonaqueous solvent, or a nonpolar side chain of a protein is buried in its hydrophobic core through folding, the transfer free energy is referred to as hydrophobic free energy. The earlier dispute concerns whether the transfer free energy can be legitimately separated into two parts and the free energy of hydrophobic hydration treated separately from the overall free energy change [3–5]. If the hydrophobic free energy is defined as the entire transfer free energy [5], then there is general agreement that transfer of the nonpolar solute (or side chain) out of water and into a nonaqueous environment drives folding in a major way. A related concern has come forward, however, and scientists increasingly question whether the energetics of forming the hydrophobic core of a protein should be attributed chiefly to packing interactions (van der Waals interactions, or dispersion forces) rather than to burial of nonpolar surface area. This question is closely related to the issue of whether the hydrophobic free energy in protein folding should be modeled by liquid–liquid transfer experiments or by gas–liquid transfer experiments. The energetic role of peptide hydrogen bonds (H-bonds) was studied as long ago as 1955 [6] but the subject has made slow progress since then, chiefly because of difficulty in determining how water interacts with the peptide group both in the unfolded and folded forms of a protein. Peptide H-bonds are likely to make a significant contribution to the energetics of folding because there are so many of them: about two-thirds of the residues in folded proteins make peptide H-bonds [7]. Peptide backbone solvation can be predicted from electrostatic algorithms but experimental measurements of peptide solvation are limited to amides as models for the peptide group. This chapter gives a brief historical introduction to the ‘‘weak interactions in protein folding’’ and then discusses current issues. It is not a comprehensive review and only selected references are given. The term ‘‘weak interaction’’ is somewhat misleading because these interactions are chiefly responsible for the folded struc127